Altered fat metabolism may help identify ITP patients: Study
Machine learning identifies 3 metabolites as potential diagnostic biomarkers
Written by |
Lipid, or fat, metabolism is altered in people with immune thrombocytopenia (ITP), resulting in distinct metabolic signatures that could help distinguish ITP patients from healthy individuals, a new study suggests.
Using untargeted metabolomics — a global analysis of all detectable small molecules, called metabolites, found in blood — researchers identified more than 1,200 metabolites that differed between people with ITP and healthy controls, many of which were related to fat metabolism.
When the team applied machine learning methods to the data, they found that blood levels of three metabolites were capable of reliably separating ITP patients from healthy individuals, highlighting their potential use as diagnostic biomarkers.
“These findings indicate that metabolomics-identified differential metabolites can feasibly support ITP classification models and may function as clinical diagnostic biomarkers,” researchers wrote, while emphasizing that larger, longitudinal studies are needed “to verify clinical translatability.”
The study, “Untargeted Metabolomic Profiling Reveals Lipid Metabolism Dysregulation in Patients With Immune Thrombocytopenia,” was published in the Scandinavian Journal of Immunology.
Fat metabolism plays important role in immune function
ITP is an autoimmune disorder in which the immune system mistakenly targets and destroys platelets, the circulating cell fragments that help blood clot, leading to the loss of platelets and an increased risk of bleeding. While immune mechanisms driving the disease have been widely studied, much less is known about how changes in the body’s metabolism contribute to ITP.
Fat metabolism plays an important role in both immune function and platelet biology. Fats are major components of cell membranes and also act as signaling molecules that regulate inflammation and immune activity. Disruptions in fat metabolism can affect both immune responses and platelet structure and function.
Previous studies have linked abnormal fat metabolism to autoimmune and inflammatory diseases such as systemic lupus erythematosus, the most common form of lupus, and multiple sclerosis, highlighting a close relationship between metabolism and immune regulation. In ITP, emerging evidence suggests that altered fat profiles may be associated with immune dysregulation and with platelet membrane abnormalities.
To better understand changes in fat metabolism in ITP, researchers in China measured thousands of metabolites simultaneously in blood samples from 20 adults with newly diagnosed ITP and compared the results with samples from 17 healthy individuals of similar age and sex. All patients required treatment due to clinically significant bleeding.
The analysis revealed a clear separation between ITP patients and healthy controls. In total, more than 1,200 metabolites differed significantly between the two groups, with 680 metabolites found at higher levels and 589 at lower levels in ITP patients.
Researchers built diagnostic prediction model
Pathway analyses showed that many of these metabolites were associated with fat metabolism pathways, particularly those involved in unsaturated fatty acid production and cholesterol metabolism. These pathways play important roles in controlling inflammation, regulating immune responses, and maintaining platelet membrane structure and function.
To assess whether these differences could help identify ITP patients, the researchers built a diagnostic prediction model using machine learning methods. This analysis identified three metabolites — 3-hydroxybutyric acid, decanoyl-sn-glycero-3-phosphocholine, and 2,2-dimethyl-N-(2,4,6-trimethoxyphenyl) dodecanamide — that best distinguished ITP patients from healthy individuals.
Levels of 3-hydroxybutyric acid were increased in ITP patients. This molecule is produced when the body shifts toward ketone metabolism, meaning it relies more heavily on fats for energy. Increased ketone metabolism has been associated with higher platelet counts and reduced inflammatory responses, suggesting that elevated levels of this metabolite may reflect disease-related metabolic changes.
These findings suggest that dysregulated lipid metabolism is a central feature of ITP and support the potential of [blood] metabolomics-based machine learning models to aid in the identification of novel biomarkers and to improve early diagnosis of this immune-mediated thrombocytopenia.
In contrast, decanoyl-sn-glycero-3-phosphocholine was found at lower levels in ITP patients. This molecule is an essential component of cell membranes. The researchers noted that platelet activation and shape changes depend on proper membrane lipid composition and fluidity, and that reduced levels of membrane-associated lipids may reflect altered platelet membrane function.
The third metabolite, 2,2-dimethyl-N-(2,4,6-trimethoxyphenyl) dodecanamide, was also detected at lower levels and linked to cholesterol metabolism. Lower levels of this compound may indicate disrupted cholesterol balance in ITP, which the researchers suggest could be relevant to abnormal clotting tendencies seen in some patients.
Using these three markers, the model correctly identified ITP patients with a sensitivity of 78% and classified healthy individuals with a specificity of 100%. The model achieved an area under the curve (AUC) of 0.83, indicating good overall ability to distinguish between people with and without the disease.
“These findings suggest that dysregulated lipid metabolism is a central feature of ITP and support the potential of [blood] metabolomics-based machine learning models to aid in the identification of novel biomarkers and to improve early diagnosis of this immune-mediated thrombocytopenia,” the scientists wrote.
